基于到发线优选罚值的高速铁路列车运行调整双层优化模型

Two-layer optimization model of train rescheduling for high-speed railway based on preferred penalty value of arrival-departure track allocation

针对不同干扰场景导致的列车晚点情况下列车运行调整计划在车站的可执行性问题,提出协同考虑列车运行调整和车站到发线运用冲突疏解的双层规划模型.上层模型以列车偏离计划运行图时刻最小为目标,考虑列车运行安全和车站到发线运用计划可行性约束,从整体上尽快恢复列车正点运行;下层模型以各车站选择到发线偏离原定计划惩罚值最小为目标,考虑到发线运用规则和安全间隔约束,为上层提供到发线运用计划可行性判断指标.同时,设计双层嵌套式遗传算法求解模型.基于武广高铁实际案例,验证所提出模型和算法的有效性.研究结果表明:设计的3种干扰场景均能在300 s内得到合理可行的列车运行调整计划及到发线分配方案,与分步调整策略和常规调整策略相比,提出的双层协同调整策略能够在提高列车运行调整质量的同时,得到各车站选择惩罚值最优的到发线运用调整方案,验证了模型及算法的有效性.

To address the issue of executability of train rescheduling plan at stations in scenarios where trains are delayed due to various interferences,this paper proposes a two-layer planning model that considers both train rescheduling and the resolution of conflicts related to arrival-departure track utiliza-tion at stations in a coordinated manner.The upper-level model aims to minimize the deviation of train schedules from the planned operation chart while considering safety and the feasibility of station arrival-departure track utilization.It focuses on restoring on-time train operations efficiently from the overall perspective.The lower-level model seeks to minimize the penalty value associated with the deviation of the arrival-departure track from the original plan for each station.It considers the operational rules of arrival-departure tracks and safety intervals,providing a feasibility assessment for the upper-level decision-making.Meanwhile,a two-layer nested genetic algorithm is designed to solve this model.The effectiveness of the proposed model and algorithm has been validated using a real-world case from the Wuhan-Guangzhou high-speed railway.The results show that under all three designed disturbance scenarios,the model obtains reasonable and feasible train recsheduling plans and arrival-departure track allocation schemes within 300 s.Compared to the stepwise adjustment strategy and the conven-tional adjustment strategy,the designed two-layer cooperative adjustment strategy not only enhances the quality of train rescheduling but also yields the optimal penalty values for arrival-departure track ad-justments at each station,confirming the efficacy of the model and algorithm.